INDEX
Explanations
short statements or phrases that end in a period
punctuation marks, particularly periods
New Auto-Interp
Negative Logits
thal
-0.66
itory
-0.63
opponent
-0.63
exclusively
-0.62
anni
-0.62
orche
-0.62
oshenko
-0.61
hing
-0.61
amn
-0.61
transition
-0.60
POSITIVE LOGITS
However
1.02
Likewise
0.96
Therefore
0.95
Secondly
0.95
Otherwise
0.95
Unfortunately
0.94
Additionally
0.92
Furthermore
0.92
Conversely
0.91
Luckily
0.90
Activations Density 0.987%